ABSTRACT Diploid organisms use many elaborate mechanisms to assess the genetic fitness of potential mating partners. Do haploid gametes make similar assessments; might they choose each other based on their genetic content? Remarkably, we know less about the genetics of fertilization than about meiosis or gametogenesis, even though the three together are the foundation of sexual reproduction. Mendel's First Law explicitly addresses the independent segregation of alleles that ensures their balanced representation among offspring. Implicit in the First Law, but rarely addressed, is the notion that the union of gametes is also random. We discovered evidence for strongly biased segregation in mice for multiple genes, on several genetic backgrounds, and in reports from several groups. But unlike known examples of bias that occur in only one sex, our examples seem to depend on the genetics of both parents. Such departures are usually attributed to reduced viability among individuals that carry the mutant allele, a conclusion that is too often claimed without direct evidence for embryo loss. Interestingly, no evidence was found for selective lethality or for segregation distortion that was an intrinsic property of either females or males. We propose that these genes cause gametes to join non-randomly at fertilization, thereby leading to segregation bias without embryo loss. We propose two tests: Aim 1 ? Preferential segregation. A key prediction of the biased-fertilization hypothesis is that meiotic divisions are irregular during oogenesis in heterozygous females. In particular, favored alleles and recombinant chromosomes are predicted to be retained in oocytes versus polar bodies, depending on maternal genetics at Meiosis I (MI) and on the genetics of fertilizing sperm at Meiosis II (MII). We will also test whether reversed meiosis, which was reported in humans after induced ovulation, occurs with similar hormonal treatment in mutant and control mice. Tests focus on Ago2 and Ppp2cb, both of affect chromosome segregation. Eggs and polar bodies from heterozygous females will be genotyped for the allele test, hybrids between mutant and wildtype strains for the recombination test, and eggs, polar bodies and embryos for the stress test. We will also test features of ovulation, spermatogenesis and embryogenesis for consistency with biased fertilization hypothesis. Aim 2 ? In vitro fertilization (IVF). We will test whether bias occurs with IVF for Ago2 and A1cf, both of which show strong and reproducible fertilization bias. Positive results will enable tests for mechanism for sperm-egg interactions given the ability to manipulate culture conditions experimentally. This work is relevant to NICHD High Priority Areas: genetics of idiopathic infertility; nutrition and reproduction.